The Sit Down logo

The Sit Down

Community
kucherenko
the-sit-down

Use when the grilling consensus is approved and it is time to draft the formal contract specification — strictly prohibits code generation, produces the binding spec document that governs all subsequent phases

Overview

Publisherkucherenko
Repositorygangsta
Skill namethe-sit-down
Stars
86
Forks
10
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by kucherenko on GitHub. Read the source before you install it.

Installation

Install the The Sit Down AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/kucherenko/gangsta/tree/master/skills/the-sit-down
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/kucherenko/gangsta.git /tmp/gangsta
mkdir -p .claude/skills
cp -r /tmp/gangsta/skills/the-sit-down .claude/skills/the-sit-down
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable The Sit Down in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads The Sit Down on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so The Sit Down is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

The Sit-Down: Planning and Spec Drafting

Overview

The Sit-Down is where the Don, Consigliere, and Underboss finalize the Contract — the binding specification for the Heist. This is a "Plan-First" phase where code generation is STRICTLY PROHIBITED.

Trigger

Invoked after the Don approves the Grilling consensus (The Grilling complete).

The Absolute Rule

NO CODE GENERATION DURING THE SIT-DOWN.

Not pseudocode. Not "example snippets." Not "draft implementations." NOTHING that looks like code. If you catch yourself writing code, STOP. You are violating Omerta Law 5.

The Contract describes WHAT and WHY. Implementation details (HOW) belong in the Work Packages created during Resource Development.

Flow

dot
digraph sit_down {
    "Synthesize inputs\n(Dossier + Grilling + Constitution)" [shape=box];
    "Propose 2-3 approaches" [shape=box];
    "Don approves approach?" [shape=diamond];
    "Draft the Contract" [shape=box];
    "Consigliere review\n(contradiction, ambiguity, completeness)" [shape=box];
    "Consigliere verdict?" [shape=diamond];
    "Don signs Contract?" [shape=diamond];
    "Invoke gangsta:resource-development" [shape=doublecircle];

    "Synthesize inputs\n(Dossier + Grilling + Constitution)" -> "Propose 2-3 approaches";
    "Propose 2-3 approaches" -> "Don approves approach?";
    "Don approves approach?" -> "Propose 2-3 approaches" [label="revise"];
    "Don approves approach?" -> "Draft the Contract" [label="approved"];
    "Draft the Contract" -> "Consigliere review\n(contradiction, ambiguity, completeness)";
    "Consigliere review\n(contradiction, ambiguity, completeness)" -> "Consigliere verdict?";
    "Consigliere verdict?" -> "Draft the Contract" [label="REJECT"];
    "Consigliere verdict?" -> "Don signs Contract?" [label="APPROVE"];
    "Don signs Contract?" -> "Draft the Contract" [label="changes requested"];
    "Don signs Contract?" -> "Invoke gangsta:resource-development" [label="signed"];
}

Process

Step 1: Synthesize Inputs

Gather and internalize all available intelligence:

  • Reconnaissance Dossier
  • Grilling Consensus or Best Available Consensus
  • Project Constitution (Commandments and Negative Constraints)
  • Relevant Ledger Entries

Step 2: Propose 2-3 Approaches

Before writing any spec, the Underboss must present the Don with 2-3 distinct viable approaches to the problem. Each approach must cover:

### Approach A: <Name>
<One-sentence description>

**Pros:** <key advantages>
**Cons:** <key trade-offs>
**Risk:** HIGH / MEDIUM / LOW
**Best when:** <conditions that favour this approach>

Rules:

  • Approaches must be genuinely distinct — not minor variations of the same idea
  • At least one conservative/safe option and one bolder option
  • No code. No pseudocode. Architecture and rationale only.

Present all approaches then ask: "Which approach do you want to proceed with, or would you like adjustments?"

Present this choice through the platform's native interactive selection tool (OpenCode question, Claude Code AskUserQuestion) — do not just dump the list as plain text and wait. Fall back to plain text only when no such tool exists.

Wait for the Don's explicit selection before writing anything.

  • Don selects / approves → proceed to Step 3
  • Don requests revision → update proposals, re-present

Step 3: Underboss Drafts the Contract

Using the approved approach as the architectural foundation, draft the formal specification.

The Contract must include:

markdown
---
heist: <heist-name>
date: YYYY-MM-DD
status: draft
signatories: []
---

# Contract: <Heist Name>

## Objective
<What is being built, in one paragraph>

## Requirements
### Functional Requirements
1. <FR-001> <Requirement>
2. <FR-002> <Requirement>
...

### Non-Functional Requirements
1. <NFR-001> <Requirement (performance, security, accessibility, etc.)>
...

## Architectural Decisions
<Key decisions from the Grilling consensus, with rationale>

## Grilling Conclusions
### Key Decisions
- <Decision>: <rationale>

### Rejected Alternatives
- <Alternative>: <why rejected>

### Unresolved Objections
- <Objection> — Risk: HIGH/MEDIUM/LOW — Mitigation: <if any>

## Applicable Constitution Rules
### Commandments
- <Commandment text> — Source: insights/<file>

### Negative Constraints
- NEVER <constraint> — Source: fails/<file>

## Acceptance Criteria
<How do we know the Heist is complete? Specific, testable criteria.>

## Out of Scope
<What this Heist explicitly does NOT cover>

## Open Risks
<From the Grilling — any unresolved objections with assessed risk>

Step 4: Consigliere Reviews

Invoke gangsta:the-consigliere for spec integrity review:

  • Contradiction scan
  • Ambiguity check
  • Completeness audit
  • Constitution alignment
  • Security review

The Consigliere returns a verdict: APPROVE, APPROVE WITH CONCERNS, or REJECT.

If REJECTED: Underboss revises based on Consigliere feedback. Re-review.

Step 5: Don Signs the Contract

Present the Contract to the Don:

"The Contract for Heist '' is ready for your review. The Consigliere's verdict: [verdict]. [If concerns, list them.] Do you approve?"

The Don may:

  • Sign — Contract is binding. Immediately invoke gangsta:resource-development — do NOT ask the Don what to do next, do NOT pause, do NOT prompt for confirmation. Auto-advance is mandatory.
  • Request changes — Underboss revises. Back to Step 4 (Consigliere re-reviews).
  • Kill the Heist — Abort. No further phases.

Update the Contract frontmatter:

yaml
status: signed
signatories: [Don, Consigliere, Underboss]

Output

Save to: docs/gangsta/<heist-name>/specs/YYYY-MM-DD-contract.md

Checkpoint

yaml
---
heist: <heist-name>
phase: the-sit-down
status: completed
timestamp: <ISO 8601>
next-action: Proceed to Resource Development
artifacts:
  - docs/gangsta/<heist-name>/specs/YYYY-MM-DD-contract.md
---

Omerta Compliance

  • Spec is Law: The Contract becomes the binding spec — all future phases must trace to it
  • Rule of Truth: All requirements cite Dossier findings or Grilling consensus
  • Rule of Availability: Contract and checkpoint saved to files
  • NO CODE: Verify zero code blocks in the Contract

Frequently asked questions

What does the The Sit Down AI skill do?

Use when the grilling consensus is approved and it is time to draft the formal contract specification — strictly prohibits code generation, produces the binding spec document that governs all subsequent phases

Why use The Sit Down on TypingMind?

Because you install it once and use it with any model. The Sit Down is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install The Sit Down in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/kucherenko/gangsta/tree/master/skills/the-sit-down. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use The Sit Down?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with The Sit Down?

As many as you like. As long as a model supports skills, you can use The Sit Down with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the The Sit Down AI skill free?

Yes. It is published on GitHub by kucherenko under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇